Ready Made Garments, Reproductive Behavior and Human Capital among Bangladeshi Women

.title[
# Ready Made Garments, Reproductive Behavior and Human Capital among Bangladeshi Women
]
.subtitle[
## SM Shihab Siddiqui
]
.author[
### 21 October, 2022
]

---

# Introduction

---
# Against the grain: Bangladeshi FLFP

- .hi[Female labor force participation (FLFP)] among 20-24 year olds stood at about 49% in 2015 (ADB 2016).

---
# Expansion of the garments industry

- Bangladeshi .hi[Ready Made Garments (RMG)] grew at about 11% a year since 1991.

.pull-left[
<img src="data:image/png;base64,#jmp_pres_files/figure-html/unnamed-chunk-2-1.svg" width="100%" style="display: block; margin: auto;" />
]

- <font size="5"> Accounts for 75-85% of Bangladesh's exports in recent decades.

]

- Contributed about 5-6% to GDP in 2019.

]

- .hi[About 60% of workers in export oriented RMG industry are women.]

</font>
]

---
# Could there be a connection?

## FLFP

- Textile and affiliated industry always employed relatively more women across different time and place.

- Mid 1800s England (Burnette, n.d), USA (Field-Hendrey, 1998); developed and developing countries 1981-2008 (Kucera and Tejani, 2014).
  
--
  
  - Women engaged in more spinning and knitting for centuries *(Virginia Postrel, Textiles and the Fabric of Civilization)*
  
--

- Did this "kick" off an overall FLFP transition in lines of Fogli and Veldkamp (2011)?

---
# Could there be a connection?

## Reproductive behavior

- Labor market opportunitieschanges net benefits of marriage and children *(e.g. Aaronson et al 2014; Greenwood et al 2017).*

- But fertility transition was already on the way and may have already ushered in a low-fertility enviroment? What about timing?

## Human capital accumulation

- More schooling if returns to schooling in RMG industry is adequate. Less if potential students are better off working.

- Maybe less schooling but more human capital accumulation through increased effort?
--

---
# Research questions

1. To what extent did the emergence of the RMG industry contribute to the increase in FLFP in Bangladesh?
  
    - A question about magnitude.
--

2. What is the effect of the RMG industry on reproductive behavior of women (marriage and fertility)?
    
    - A question about sign and magnitude.
  
--

3.  What is the effect of the RMG industry on human capital accumulation of Bangladeshi women?
  
    - A question about sign and magnitude.

---
# Why do we care?

1. Adds to the literature on manufacturing- and export-led growth. Especially relevant since:

--
  - General concerns with pre-mature industrialization (Rodrik, 2015)
  
  - Reductions in prevalence of women in RMG industry in Bangladesh <footnote>A [Along the lines of what happens as technology improves in a manufacturing sector (Tejani and Kucera, 2021).</footnote>

2. Adds to the literature focusing on trade and lives of workers (Autor et al (2013), Li (2018) and Autor et al (2019)).

---
# Preview of the paper

## Methods

- Estimates the long run impact of female labor demand shock on FLFP, fertility and human capital accumulation by:

- Bartik shift-share method to identify labor demand shocks following a methodology similar to .hi[Autor et al (2013).]
  
  - Specifically, I exploit product specialization in the RMG industry along the knit versus woven line across sub-districts (Bangladesh administrative level-3) for identification.

---
# Preview of the paper

## Results

- FLFP, particularly industrial FLFP changes a lot.

- Not much of an impact on reproductive behavior and fertility overall.

-- Results can be rationalized with previous literature under plausible assumptions.

---

# RMG industry in Bangladesh

---
# Knit versus Woven prodcuts

- Single yarn looped repeatedly.
- HS code 61.
- Product examples: Most sweaters, cotton T-shirts.
]

- Multiple yarn criss-crossed over and under each other.
- HS code 62.
- Product examples: Shirts, jackets, pants.
]
---
# Knit versus Woven Specialization

- Producing woven is more energy and capital intensive, and commands about 10% higher per unit price (Sytsma, 2022).

- Woven factories are larger, and employs more women.

---
# Location of RMG factories

- RMG factory location choice is mostly dependent on infrastructure quality (kagy, 2014).

---
count:false

# Identification strategy and data

---
# Overview of the identification strategy

- Sub-districts with and without factories are likely to have different infrastructure quality, which maybe correlated with outcome variables. So, .hi[I restrict the analysis only to sub-districts that had a factory by 2006.]
  
--

- Different sub-districts have different intensity of knit versus woven specialization **within** the RMG industry.

--
    - Adopting Goldsmith-Pinkham et al (2020), the key assumption is that these differences in specialization do not change outcomes through confounders.
 
---
# Regression Model

`\begin{equation}
  \Delta Y_{s,t} = \beta \space \Delta \text{Export Exposure}_{s,t} + \delta_{t}  + Z_{s,t-1} \beta_z + X_{s,t-1} \beta_x + \epsilon_{s,t}
\end{equation}`

- `\(\Delta Y_{s,t}\)` is the decadal change in outcome variables in sub-district `\(s\)` over decade ending at year `\(t\)`.

- `\(\text{Export Exposure}_{s,t}\)` measures export exposure in sub-district over decade ending at year `\(t\)`.

- `\(\delta_t\)` are period fixed-effects and `\(Z_{s, t-1}\)` is a vector of controls across all outcomes including start of period electrification rate, urbanization, density, share of 15-64 year old in population and years of education for adults (15-64).

- `\(X_{s,t-1}\)` are start of the period comparable outcomes for males other than in the cases of regressions corresponding to marriage and fertility rates.

---
# Export exposure

One candidate measure of export exposure per potential worker (Population 15-64) is:

`\begin{equation}
  \sum_{i=0}^{9} \alpha_{s,t-i}^{K} * \frac{\text{Export}_{BD,t-i}^{K}}{L_{s,t-i}} + \sum_{i=1}^{9} \alpha_{s,t-i}^{W} * \frac{\text{Export}_{BD,t-i}^{W}}{{L_{s,t-i}}} \\
 \alpha_{s,t-i}^{K} = \frac{Machines_{s,t-i}^{K}}{Machines_{BD,t-i}^{K}}, 
\alpha_{s,t-i}^{W} = \frac{Machines_{s,t-i}^{W}}{Machines_{BD,t-i}^{W}}
\end{equation}`

- Apportions total knit (woven) exports originating in Bangladesh to a sub-districts based on what share of national knit (woven) production is in the sub-district.

- Export exposure in sub-district `\(s\)` in at time period `\(t\)` depends on the intensity of knit (woven) specialization and is scaled by total exports and population.
  
---
# Export exposure

- But clearly the previous measure can be endogenous since over decades, infrastructure changes and that in itself could change both the shares and change the outcomes differentially. So, I fix share to the values at the start of decade. Thus export exposure per potential worker is as follows:

`\begin{align}
  \Delta \text{ Export Exposure}_{s,t} =& \alpha_{s,t-1}^{K} * \frac{\Delta \space \text{Export}_{BD,t}^{K}}{L_{t-1}} + \alpha_{s,t-1}^{W} * \frac{\Delta \space \text{Export}_{BD,t}^{W}}{L_{t-1}} 
\end{align}`

---
# Identifying Assumptions

Adopting from Goldsmith-Pinkham et al (2020), the key assumption is that the differences in knit versus woven specialization do not *change* outcomes through confounders. That is:

- .hi[Identifying assumption 1:] Extent of knit versus woven specialization in a sub-district is uncorrelated with the errors conditional on controls in the first difference equations.

- .hi[Identifying assumption 2:] FLFP responds similarly to woven and knit shocks.

- .hi[Example of a violation:]	Woven employs more women, so areas with more woven factories has increased presence of fertility control programs because there are more women.

---
# Data Sources

- .hi[Outcome data] is obtained by aggregating individual-level data from the Bangladesh Census 1991 (10% sub-sample), 2001 (10% sub-sample) and 2011 (5% sub-sample).

- .hi[Factory data] are primarily from  multiple BGMEA datasets from members directory (2001, 2010), BGMEA (2015) and scraping of BGMEA websie in 2013.

---
class: inverse, middle, center

# Results

---
# FLFP

## Overall

<table class="kable_wrapper">
<caption>Influence on overall (left) and industrial (right) FLFP</caption>
<tbody>
  <tr>
   <td>

<table>
 <thead>
  <tr>
   <th style="text-align:center;"> Age </th>
   <th style="text-align:center;"> coeff </th>
   <th style="text-align:center;"> std_error </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:center;"> 15-64 </td>
   <td style="text-align:center;"> 0.004 </td>
   <td style="text-align:center;"> 6e-04 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 15-29 </td>
   <td style="text-align:center;"> 0.005 </td>
   <td style="text-align:center;"> 8e-04 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 15-20 </td>
   <td style="text-align:center;"> 0.006 </td>
   <td style="text-align:center;"> 9e-04 </td>
  </tr>
</tbody>
</table>

</td>
   <td>

<table>
 <thead>
  <tr>
   <th style="text-align:center;"> Age </th>
   <th style="text-align:center;"> coeff </th>
   <th style="text-align:center;"> std_error </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:center;"> 15-64 </td>
   <td style="text-align:center;"> 0.007 </td>
   <td style="text-align:center;"> 8e-04 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 15-29 </td>
   <td style="text-align:center;"> 0.006 </td>
   <td style="text-align:center;"> 6e-04 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 15-20 </td>
   <td style="text-align:center;"> 0.007 </td>
   <td style="text-align:center;"> 7e-04 </td>
  </tr>
</tbody>
</table>

</td>
  </tr>
</tbody>
</table>

- Modest impact on FLFP that is stronger on industrial FLFP and among younger women.

--
- At mean export exposure for a sub-district over two decades:

- FLFP increases by 3.45 percentage point overall, 6.04 for industrial FLFP.
  
- More than half of industrial FLFP change is from RMG industry.
  
---
# Reproductive behavior

<table class="kable_wrapper">
<caption>Influence on marriage rates (left) and fertility (right)</caption>
<tbody>
  <tr>
   <td>

<table>
 <thead>
  <tr>
   <th style="text-align:center;"> Age </th>
   <th style="text-align:center;"> coeff </th>
   <th style="text-align:center;"> std_error </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:center;"> 15-20 </td>
   <td style="text-align:center;"> -4e-04 </td>
   <td style="text-align:center;"> 5e-04 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 21-30 </td>
   <td style="text-align:center;"> 1e-04 </td>
   <td style="text-align:center;"> 4e-04 </td>
  </tr>
</tbody>
</table>

</td>
   <td>

<table>
 <thead>
  <tr>
   <th style="text-align:center;"> Age </th>
   <th style="text-align:center;"> coeff </th>
   <th style="text-align:center;"> std_error </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:center;"> 15-20 </td>
   <td style="text-align:center;"> 0.0001 </td>
   <td style="text-align:center;"> 0.0004 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 21-30 </td>
   <td style="text-align:center;"> 0.0000 </td>
   <td style="text-align:center;"> 0.0015 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 15-20 </td>
   <td style="text-align:center;"> 0.0036 </td>
   <td style="text-align:center;"> 0.0023 </td>
  </tr>
</tbody>
</table>

</td>
  </tr>
</tbody>
</table>

- Signs as I expected, but not statistically significant.

---
# Human capital accumulation

<table>
<caption>Influence on schooling rates of women</caption>
 <thead>
  <tr>
   <th style="text-align:center;"> Age </th>
   <th style="text-align:center;"> coeff </th>
   <th style="text-align:center;"> std_error </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:center;"> 05-09 </td>
   <td style="text-align:center;"> 0.0009 </td>
   <td style="text-align:center;"> 0.0008 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 10-13 </td>
   <td style="text-align:center;"> -0.0002 </td>
   <td style="text-align:center;"> 0.0008 </td>
  </tr>
  <tr>
   <td style="text-align:center;"> 14-19 </td>
   <td style="text-align:center;"> -0.0013 </td>
   <td style="text-align:center;"> 0.0011 </td>
  </tr>
</tbody>
</table>

- Signs as I expected, but not statistically significant.
---
# Conclusions

- Results indicate that the RMG industry provided opportunities for some of the women to join manufacturing labor force.

- However, most changes in reproductive and human capital likely driven by other factors.
--